In order to summarise the findings of this study, the descriptive and qualitative
data as well as the solutions to the specific research questions under investigation are
discussed in terms of the four review questions originally presented in the literature
Chapter Six
with PCA?, and (6.1.4) does control influence pain management in PCA? This section
ends with (6.1.5) a summary of the findings.
6.1.1 Is PCA Effective?
The findings of previous research have concluded that compared to traditional
intramuscular injections, PCA is an effective method of providing post-operative
analgesia (Ballantyne et al., 1993; Thomas et al., 1995; W asylak et al., 1990). The
present study did not compare PCA with other methods of post-operative pain relief,
but participants’ responses to the 17 items of the PCA satisfaction questionnaire
contribute to the understanding of why PCA has been effective. As presented in Table
3, the vast majority of participants reported that PCA did not provide poor overall pain
relief. In addition, the vast majority of participants reported that PCA gave effective
relief when resting, while only half of participants responded that PCA provided
effective pain relief when moving or coughing. It is interesting that more than two
thirds of participants responded that PCA worked quickly, while just over half of the
participants reported that there was pain after surgery before the PCA became
effective. These findings suggest that for the participants in this study, PCA provided
successful pain relief overall and when they were at rest. However, only half of the
participants found PCA to be effective for pain relief when they were moving or
coughing. Although more than half of the women assessed in this study experienced
pain after surgery before PCA became effective, the majority found that PCA began to
work quickly. While these findings add to the understanding of PCA effectiveness, it
Chapter Six
PCA use. The results of research question one that tested the impact of nausea and
vomiting as side effects suggested that there was a statistical difference that
approached significance between the mean total volume of analgesia consumed during
PCA use for participants with and without post-operative side effects. Although this
difference was not statistically significant, it is of clinical importance. For PCA to be
a consistently effective method of providing post-operative acute pain relief, patient
blood levels of the opioid drug morphine should be kept within the analgesic corridor
where pain relief is achieved without the experience of side effects. Previous research
has found PCA to be effective for post-operative pain, but additional research is
needed to help determine specific reasons for its effectiveness.
6.1.2 Is Patient Satisfaction with PCA Important?
Pain relief is an essential element of post-operative patient care. However, beyond
the individual differences in analgesic drugs needed to achieve pain relief, patient
satisfaction with PCA has been described as an important theme. The results of
research question one from this study suggest that PCA satisfaction scores were a
statistically significant predictor variable of membership in two groups. The first
group consisting of 20 participants who reported post-operative side effects had
significantly lower PCA satisfaction scores compared to the 25 participants in the
second group who did not report side effects. This finding supported the hypothesis
generated by the research of Jamison et al. (1993) that post-operative side effects
would reduce participant’s satisfaction with PCA. This quantitative result concurred
Chapter Six
vomiting were important complications of PCA satisfaction.
Research question four with PCA satisfaction scores as the criterion variable in a
multiple regression analysis failed to partially replicate the findings of Jamison et al.
(1993). These researchers concluded that PCA satisfaction scores measured with a 7-
point Likert-type scale were predicted by preoperative anxiety and post-operative pain
intensity. The non-significant result for research question four suggests that the
preoperative and post-operative variables as well as volume of morphine consumed by
participants did not help to predict the variance in PCA satisfaction scores as measured
in this research. Although PCA satisfaction scores helped to significantly discriminate
participants with and without side effects, the present results are somewhat
inconclusive because they were unable to empirically identify predictors of PCA
satisfaction. In addition, only one of the participants who provided preoperative
qualitative information mentioned PCA and satisfaction. The only specific conclusion
generated from research question four of this study is that PCA satisfaction is a
complex concept.
The idea forwarded by Egan and Ready (1994) that successful pain relief may
intuitively contribute to PCA satisfaction was not supported by the results of the
present research. PCA satisfaction scores did not significantly predict post-operative
pain in the regression analyses of research question two. Moreover, pain intensity and
the log of pain quality scores failed to significantly predict PCA satisfaction scores in
the regression analysis of research question four. The responses provided by
participants to the 17 items of the PCA satisfaction questionnaire designed by Egan
Chapter Six
understand the complex concept of satisfaction with PCA. In a paper published just
this year, Chumbley, Hall and Salmon (1999) attempted to go beyond measuring the
concept of PCA satisfaction by identifying what they referred to as aspects of patient’s
experience with PCA. These authors chose not to ask patients about their satisfaction
with PCA largely because of the complexity of the concept, and because patients’
answers can include evaluations of aspects of their care unrelated to PCA. In short,
the validity of the concept of patient satisfaction appears to be in question (Williams,
1994). Chumbley et al. (1999) concluded that scores of patient satisfaction are often
used to evaluate the adequacy of treatment, but unfortunately they may be an
insensitive index of patients’ views. Once again, the findings of the present study in
relation to PCA satisfaction are somewhat inconclusive. A discussion of the present
limitations as well as directions for future research with regards to this concept are
presented in sections 6.4 and 6.5 of this chapter.
6.1.3 Can Anxietv Affect Pain Management with PCA?
In contrast to the findings of Gill et al. (1990), Perry et al. (1994) and Thomas et
al. (1995), there was no evidence within the present study to suggest that state anxiety
was a significant predictor of post-operative pain. State anxiety scores did not emerge
as a statistically significant predictor variable within the results of the two separate
regression analyses for research question two with pain intensity scores and the log of
pain qualities index as criterion variables. The qualitative data provided before surgery
was consistent with these quantitative findings, as only 3 (11%) participants mentioned
Chapter Six
study to partially replicate previous findings may be due to differences in post
operative pain scores. Thomas et al. (1995) reported a mean of 21.6 for the qualities
of pain index of the M PQ-SF for their PCA participants, while in the present study
prior to being transformed due to a significant degree of kurtosis the mean qualities of
pain index score was 10.91 for the 45 participants. The mean scores for the Thomas
et al. (1995) study was the average of post-operative pain assessments conducted at 6,
18 and 24 hours after surgery, while post-operative pain in the present study was
measured only once 24-36 hours after surgery. The multidimensional nature of pain
and the lack of consistency with procedures employed in its measurement may help
explain the failure with this study to partially replicate the previous findings that
anxiety is associated with post-operative pain.
The results for research question three do, however, partially replicate the findings
of Thomas et al. (1995) that state anxiety scores help predict total volume of morphine
consumed by participants. The entry of state anxiety and emotional distress in step 3
of this regression equation made a significant contribution to the prediction of total
volume of morphine consumed, following the entry of dimensions of pain in step 1
and self-efficacy beliefs in step 2. State anxiety as measured by the STAI helped to
predict 11% of the variance in morphine volume consumed by the participants. This
was, however, lower than the 23% of variance accounted for by Thomas et al., (1995).
This difference may in part be explained by the inclusion of both IMI and PCA
participants in the Thomas et al. (1995) study. An additional explanation may be
related to differences in the total volume of morphine consumed. In the present study,
Chapter Six
This was considerably lower than the two separate means reported by Thomas et al.
(1995). Their PCA participants consumed on average 53mg of morphine, while their
IM I participants consumed on average 79mg of morphine.
The mean total STAI state anxiety score for the 45 participants in the present
study was 41.96, which was similar to mean of 44.3 reported by Thomas et al. (1995)
for their 110 participants. When compared to the normative STAI state anxiety score
of 36 for women between 19 and 49 years of age provided by Spielberger et al.
(1983), the present sample of hysterectomy patients had higher scores but were within
the normal range. The potentially stressful aspects of hospital admission for surgery
may have caused this increase in state anxiety of the participants. Heath and Thomas
(1993) have reported that moderate levels of preoperative anxiety are associated with
optimal recovery from surgery. As reduced pain is one of the goals of effective post
operative patient care and an important element of recovery, the results of the present
study cannot conclude that anxiety predicts post-operative pain. A conclusion that can,
however, be derived from the result of research question three is that greater
preoperative state anxiety levels were predictive of higher post-operative morphine
consumption. Munafo (1998) argues that failure to demonstrate a clear relationship
between anxiety and post-operative pain is due to an inadequate conception of anxiety
as a homogeneous construct. If anxiety is composed of affective, cognitive and
behaviour!al components as Munafo (1998) suggests, it may be possible for anxiety to
produce behaviours such as PCA dose activation in an attempt to respond to pain.
Distinguishing the components of anxiety is difficult, but the behaviour!al aspects of
Chapter Six
6.1.4 Does Control Influence Pain with PCA?
Offering patients some control over their post-operative pain in the form of an
analgesia delivery system such as PCA can potentially influence their recovery from
surgery (Heath and Thomas, 1993; Welchew, 1995). Only 4 participants mentioned
directly the importance of control within the qualitative findings of this study.
Although this information was gathered prior to surgery and therefore before PCA use,
it is consistent with the post-operative qualitative data reported by Taylor et al. (1996).
In contrast to this was the finding presented in Table 3, that after using it all but one
of the 45 participants responded that PCA provided personal control. This finding is
consistent with those of Egan and Ready (1994) who reported that over 80% of
participants found PCA provided personal control.
The role played by maladaptive cognitive coping strategies in the present study
were emphasised in the results of research question one and two. Maladaptive coping
strategies were one of two significant variables that helped discriminate participants
with from those without post-operative side effects in research question one. This
indicated that those who experienced post-operative side effects were more likely to
employed maladaptive strategies to cope with their acute pain. It is not suggested here
that the use of maladaptive strategies helped to cause post-operative side effects, but
that these ways of thinking about pain were more common in participants who
reported analgesic side effects. The first regression analysis of research question two
established that maladaptive cognitive coping strategies made a unique contribution to
the prediction of pain intensity scores. This is consistent with the conclusions of Pick
Chapter Six
maladaptive thoughts. Although the relation between pain intensity and maladaptive
coping responses in acute pain is unclear, an alternative explanation was that negative
(maladaptive) thoughts may intensify the experience of pain following a hysterectomy.
6.1.5 A Summary o f the Findings
The concept of self-efficacy beliefs as measured in this study failed to emerge as a
significant predictor variable in the analyses of post-operative pain, volume of
morphine consumed, and PCA satisfaction. These results are unable to empirically
support the conclusion of Skevington (1996) that self-efficacy beliefs affect acute pain
experiences and that self-efficacy beliefs have implications for pain management with
PCA. As a core element of Bandura’s social-cognitive theory, self-efficacy beliefs
have not contributed to the understanding of pain management with PCA in this study.
There was empirical support for the hypothesis generated by Jamison et al. (1993)
that those who experience post-operative side effects were less satisfied with PCA and
employed negative or maladaptive cognitive methods to cope with pain. It was
speculated that they possibly relied too much on their analgesia, which caused nausea
and vomiting due to high blood concentrations. Maladaptive cognitive coping
responses was also a significant predictor of pain intensity scores. As a partial
replication of the findings of Thomas et al. (1995), state anxiety scores measured prior
to surgery were a significant predictor of the total volume of morphine consumed by
Chapter Six
6.2 Clinical Implications
With the trend towards providing cost effective patient care and efficient delivery
of medical services, the general aim of this research was to further understand the
psychological factors which make PCA beneficial in order to help allocate this method
of post-operative pain relief. The presents results failed to demonstrate that the
preoperative variables including self-efficacy beliefs, emotional distress, state anxiety
and pain expectations helped to predict participants post-operative pain and PCA
satisfaction. Therefore, no evidence is available from this study to help determine
what patients would find PCA an optimal method of pain relief prior to its use. There
is evidence, however, to support the use of psychological interventions focused on
preoperative state anxiety and responses used during the pain experience as coping
strategies. Education for patients undergoing surgery can enhance the effectiveness of
their treatment (Mahler & Kulik, 1990). More specifically, Thomas et al. (1995) have
concluded that anxiety and poor coping skills can be improved by preoperative
education, which in turn may influence their recovery. From the qualitative findings
of the present study it appears that some women did attempt to prepare for their
hysterectomy by talking with others, reading about the procedure and recovery, as well
as physically preparing for surgery. The impact of these personal forms of preparation
is unclear, but formal education programmes designed to advise surgical patients about
post-operative recovery have been successfully used.
Preoperative education may have a number of clinical effects. Education can
provide reassurance with regards to patients’ fears concerning PCA use including the
Chapter Six
improve the management of analgesic side effects. Post-operative nausea and vomiting
are expensive for the health service in terms of length of time in hospital for patients
to recover (Tramer, Phillips, Reynolds, McQuay, & Moore, 1999). Education can be
used to reduce patient anxiety, encourage the autonomy and control that PCA allows,
and increase patients’ ability to actively participate in recovery (Griffin, Brennan, &
McShane, 1998; Scott & Hodson, 1997). To achieve these effects, intensive
preoperative tuition that includes an overview of the use of PCA, and the potential
impact of anxiety and stress on recovery has been recommended (Griffin et al., 1998).
M oreover, education about adaptive cognitive coping strategies and relaxation
techniques for use during the immediate post-operative period have also been
recommended (Scott & Hodson, 1997; Symonds, 1998). Relaxation techniques from a
cognitive-behaviourial framework have been used successfully to help patients cope
more effectively with a surgical situation (Laframboise, 1989). In addition to
relaxation, cognitive methods such as attention control, dissociation and distraction
strategies could be presented to patients to help them respond in an adaptive manner to
their acute pain.
Information sheets and educational videos could be employed on surgical wards to
educate patients upon admission. Although PCA information sheets are often
provided to patients, it is essential that they contain specific details o f the PCA to be
used. Based upon the ideas of Griffin et al. (1998), PCA information sheets should
include the following sections: (i) How to use PCA, with a summary o f dose size and
the lock-out interval, (ii) You’re in control with PCA, (iii) How to deal with sudden or
Chapter Six
healing and rapid recovery, (vi) W hat makes PCA safe, and (vii) Side effects, with a
summary of the analgesic corridor. Nursing staff on surgical wards are instrumental in
the implementation of preoperative information. Their knowledge of PCA use, in
addition to their skills in the assessment of patient anxiety and post-operative coping
responses can be beneficial in terms of patient recovery (Rundshagen, Schnabel,
Standi, & Schulte, 1999). The clinical implications presented may have an effect
upon levels of anxiety and post-operative coping skills that help reduce pain and
discomfort, but it is not known how these influence patient satisfaction with PCA.